List your company

Text-to-Image Retrieval on RSICD (test)

15.59R@1GeoRSCLIP-FT
2.30925.75719.20512.6529Jun 20, 2023Dec 22, 2023Jun 25, 2024Dec 27, 2024Jul 1, 2025Jan 2, 2026Jul 7, 2026

Evaluation results

MethodLinks
GeoRSCLIP-FTParadigm=GVLM + FT, Tuned on=RS5M + RET-22023.0615.5941.1957.9938.87—
GeoRSCLIP-FTParadigm=GVLM + FT, Tuned on=RS5M + RSICD2023.0615.2640.4657.7938—
ELC0Backbone=ViT-B/322026.0714.0337.3754.38—213.47
ELCBackbone=ViT-B/322026.0714.0338.0154.71—215.2
RemoteCLIPParadigm=GVLM + FT, Tuned on=RET-3 + DET-10 + SEG-42023.0613.7137.1154.2535.26—
RemoteCLIPBackbone=ViT-B/322026.0713.7137.1154.25—211.57
FGVLABackbone=ViT-B/322026.0712.7236.6153.76—204.74
PAUBackbone=ViT-B/322026.0712.393553.47—198.39
AIRBackbone=ViT-B/322026.0712.1937.0753.94—199.46
InfoNCEBackbone=ViT-B/322026.0711.8935.3153.14—196.49
GLGSABackbone=ViT-B/322026.0711.7534.0750.12—192.4
PE-RSITRParadigm=GVLM + FT, Tuned on=RSICD2023.0611.6333.9250.7331.12—
PE-RSITRBackbone=ViT-B/322026.0711.6333.9250.73—186.7
GeoAlignCLIPBackbone=ViT-L/142026.0311.431.2944.34——
FGCLIPBackbone=ViT-L/14-3362026.0311.1829.8343.31——
VSE∞Backbone=ViT-B/322026.0711.1832.8347.7—193.08
VSE++Backbone=ViT-B/322026.0710.7430.4145.56—185.35
PolyBackbone=ViT-B/322026.0710.4831.7546.18—181.18
GeoRSCLIPBackbone=ViT-L/14-3362026.0310.3628.4442.09——
LRSCLIPBackbone=ViT-L/142026.0310.0128.643.73——
GeoRSCLIPBackbone=ViT-L/142026.039.9228.0142.09——
CUPBackbone=ViT-B/322026.079.8631.647.15—182.02
GeoAlignCLIPBackbone=ViT-B/162026.039.5528.0741.92——
GeoRSCLIPParadigm=GVLM + FT, Tuned on=RS5M2023.069.5227.3740.9926.18—
LRSCLIPBackbone=ViT-B/162026.039.1527.4339.65——
MultilanguageParadigm=Supervised, Tuned on=RSICD2023.069.1428.9644.5927.42—
SIRSBackbone=ViT-B/322026.078.7430.3849.5—166.31
MTGFEParadigm=Supervised, Tuned on=RSICD2023.068.6727.5643.9230.68—
FGCLIPBackbone=ViT-B/162026.038.625.3637.62——
GeoRSCLIPBackbone=ViT-B/322026.038.4525.0138.35——
GeoRSCLIPParadigm=GVLM + Pfeiffer, Tuned on=RS5M2023.068.1224.8138.7923.94—
FAAMIParadigm=Supervised, Tuned on=RSICD2023.068.1125.5941.3723.18—
SMLGN2026.077.8527.1442.58—149.21
KAMCL2026.077.7825.2340.02—150.52
GeoRSCLIPParadigm=GVLM + UniPELT, Tuned on=RS5M2023.067.5223.3535.6622.39—
LongCLIPBackbone=ViT-L/142026.037.1423.4835.55——
SkyCLIPBackbone=ViT-B/322026.036.9921.6533.65——
PIRParadigm=Supervised, Tuned on=RSICD2023.066.9724.5638.9224.46—
GeoRSCLIPParadigm=GVLM + LoRA, Tuned on=RS5M2023.066.7323.2435.5921.33—
LongCLIPBackbone=ViT-B/162026.036.6820.6832.17——
SkyCLIPBackbone=ViT-L/14-pct302026.036.420.1532.86——
DOVEParms=38 M2023.106.0423.9540.35——
SkyCLIPBackbone=ViT-L/14-pct502026.036.0220.3132——
GeoRSCLIPParadigm=GVLM + Prefixtuning, Tuned on=RS5M2023.065.9520.3732.9919.69—
DOVE-SParms=25 M2023.105.8421.8736.54——
CLIP-BaselineParadigm=GVLM, Tuned on=None2023.065.7817.7327.7615.74—
SWANParadigm=Supervised, Tuned on=RSICD2023.065.5622.2637.4120.61—
SWANParms=40 M2023.105.5622.2637.41——
HVSAParadigm=Supervised, Tuned on=RSICD2023.065.5121.1334.1320.16—
HVSA2023.105.5121.1334.13——
CLIPBackbone=ViT-B/322026.075.4717.5327.78—93.96
KCR2023.105.422.4437.36——
CLIPBackbone=ViT-B/162026.035.2916.9626.77——
CLIPBackbone=ViT-L/142026.034.9618.7229.66——
AFMFNParadigm=Supervised, Tuned on=RSICD2023.064.918.2831.4416.42—
AMFMN2026.074.918.2831.44—98.49
KCRParadigm=Supervised, Tuned on=RSICD2023.064.7618.5927.219.14—
GaLRParadigm=Supervised, Tuned on=RSICD2023.064.6919.4832.1318.96—
GaLRParms=46 M2023.104.6919.4832.13——
GaLR2026.074.6919.4832.13—113.78
LW-MCRParadigm=Supervised, Tuned on=RSICD2023.064.6617.5130.0214.66—
LW-MCR2023.104.6617.5130.02——
VSE++Parms=29 M2023.104.3715.3725.35——
CAMPParms=63 M2023.104.2515.8227.82——
AMFMNParms=36 M2023.104.081730.6——
CAMERAParms=64 M2023.10415.9326.97——
SCAN i2tParms=60 M2023.103.9315.225.53——
SCAN t2iParms=60 M2023.103.8215.728.28——
VSE++Paradigm=Supervised, Tuned on=RSICD2023.062.8211.3218.110.43—